Importance:In a phase 2 randomized clinical trial, high-dose vitamin D3 added to standard treatment improved progression-free survival (PFS) compared with standard-dose vitamin D3 in patients with metastatic colorectal cancer (mCRC). Objective:To determine if high-dose vitamin D3 added to standard chemotherapy improves outcomes in patients with previously untreated mCRC. Design, Setting, and Participants:Double-blind phase 3 randomized clinical trial enrolling 455 patients with previously untreated mCRC, conducted in the US through the National Clinical Trials Network from October 2019 to December 2022 (database freeze: July 15, 2024). Interventions:mFOLFOX6 (modified FOLFOX6 [5-fluorouracil, leucovorin, oxaliplatin]) or FOLFIRI (5-fluorouracil, leucovorin, irinotecan) plus bevacizumab every 2 weeks with either high-dose vitamin D3 (8000 IU daily × 14 days as loading dose followed by 4000 IU daily) or standard-dose vitamin D3 (400 IU daily) until disease progression, intolerable toxicity, or withdrawal of consent. Main Outcomes and Measures:The primary end point was PFS assessed by the unstratified log-rank test. Secondary end points included objective response rate, overall survival, and toxicity. Prespecified subgroup analyses of PFS were performed according to known prognostic factors. Results:Among 455 randomized patients (median age, 59 years; 181 [40%] female) with median follow-up 20 months, the median PFS for high-dose vitamin D3 (n = 228) was 11.8 months (95% CI, 10.3-13.3) vs 10.3 months (95% CI, 9.4-12.2) for standard-dose vitamin D3 (n = 227) (1-sided log-rank P = .25). There were no significant differences in objective response rate between high-dose and standard-dose vitamin D3 (51% [95% CI, 44%-58%] vs 44% [95% CI, 37%-50%], respectively; P = .12), or in overall survival (median, 25.6 vs 27.0 months; 1-sided log-rank P = .66). There were no clinically meaningful differences in the most common grade 3 or greater adverse events between the high- and standard-dose groups, including neutropenia (n = 67 [32%] vs n = 62 [30%]) and hypertension (n = 42 [20%] vs n = 49 [23%]) or in incidence of vitamin D-associated toxicities. Conclusions and Relevance:Among patients with previously untreated mCRC, addition of high-dose vitamin D3, vs standard-dose vitamin D3, to standard chemotherapy plus bevacizumab did not improve PFS. Trial Registration:ClinicalTrials.gov Identifier: NCT04094688.

¹⁸F-fluorodeoxyglucose positron emission tomography/computed tomography (¹⁸F-FDG PET/CT) has become a cornerstone in the management of head and neck squamous cell carcinoma (HNSCC). While conventional imaging defines anatomical extent, PET provides complementary metabolic information crucial for staging, prognostication, treatment adaptation, and surveillance. This review aims to synthesize current evidence on the role of PET imaging in HNSCC, integrating recent data on quantitative biomarkers, novel radiotracers, and artificial intelligence (AI)–based analytical approaches. PET/CT demonstrates superior sensitivity for nodal and distant metastases compared with CT or MRI, particularly in advanced-stage or unknown-primary disease. Quantitative metrics such as SUVmax, metabolic tumor volume, and total lesion glycolysis provide independent prognostic value beyond TNM staging. In the post-treatment setting, ¹⁸F-FDG PET/CT reliably identifies complete metabolic response, reducing unnecessary neck dissections. Limitations include false positives early after therapy due to inflammation or osteoradionecrosis; dual-timepoint acquisition and PET/MRI may enhance specificity. New tracers targeting hypoxia (¹⁸F-FMISO), proliferation (¹⁸F-FLT), angiogenesis (⁶⁸Ga-RGD), and cancer-associated fibroblasts (⁶⁸Ga-FAPI) show promise for biological target delineation. AI methods, including radiomics and deep learning, have demonstrated potential for automated nodal detection and outcome prediction, though clinical validation remains limited. ¹⁸F-FDG PET/CT, complemented by emerging molecular tracers and AI-driven analysis, represents a pivotal tool for personalized management of HNSCC. By integrating anatomical, metabolic, and biological information, PET imaging supports refined staging, individualized radiotherapy planning, and improved patient outcomes. PET/CT improves post-treatment response assessment in HNSCC. High NPV supports PET/CT in surveillance strategies. PET/CT enhances the detection of regional and distant recurrences. Imaging timing post-CRT is key for diagnostic accuracy. PET/CT may guide personalized follow-up and salvage therapy.

Objectives:There is a growing need to develop user-friendly, bladder-specific image analysis tools that can produce reliable artificial intelligence (AI)-quantitative imaging biomarkers (QIBs) derived from multiparametric (mp)MRI data for clinical applications. To address it, we developed an AI-powered BLADdEr multiparametric MRI Analysis for Clinical Application (AI-BLADE, current release v1.0) toolbox designed for extracting mpMRI-derived quantitative metrics. Methods:AI-BLADE is an advanced tool for bladder-specific mpMRI data analysis with 2 core functionalities: (1) Deep Feature Analysis (MRI-DFA toolkit) and (2) Data-Driven Model-Based Analysis (MRI-MBA toolkit). AI-BLADE offers customizable options and serves as a one-stop shop solution for bladder cancer (BCa) clinical applications. The models within DFA and MBA were tested separately on 2 patient cohorts. DFA was used to classify BCa histology subtypes (n = 104) with T2-weighted images, while MBA was used to interrogate tumour physiology by deriving mpMRI QIBs, including apparent diffusion coefficient (ADC), and volume transfer constant (Ktrans) obtained from 34 BCa patients. Results:Out of the 17 AI models tested, the VGG19 model with a decision tree classifier and no feature selection for the fully connected layer 7 achieved the highest area under the curve of the receiver operating characteristic of 0.79 in classifying BCa histology subtypes, demonstrating the strongest performance. The mean ADC and Ktrans values were 1.22 × 10-3 (mm2/s) and 0.27 (min-1), respectively, reflecting underlying tumour physiology. Conclusion:The AI-BLADE (v1.0), a flexible and user-friendly software toolbox for analysing mpMRI data, shows strong potential for application in BCa oncology, offering capabilities that can enhance diagnostic accuracy and support improved patient outcomes. Advances in knowledge:This is the first study to design, develop, and implement a novel bladder-specific AI toolbox for analysing mpMRI data. AI-BLADE enables an advanced image analysis workflow, facilitating AI-QIB-based clinical decision-making for patients with BCa.

This study aimed to investigate the application of T2-based MRI delta-radiomics as a novel predictive tool for neoadjuvant chemotherapy (NACT) response in patients with osteosarcoma. We retrospectively analyzed data from 152 patients with pathologically confirmed osteosarcoma who underwent NACT at our institution. Axial T2-weighted MRI sequences were acquired both at baseline (pre-NACT) and after NACT (post-NACT). After image segmentation and preprocessing, 1158 radiomic features were extracted from the T2-weighted images. We developed and compared four models: the conventional quantitative imaging features-based model (CQIF model), the pre-NACT radiomics model, the post-NACT radiomics model, and the Delta-Radiomics model. Model performance was assessed using the area under the receiver operating characteristic curve (AUC) and accuracy (ACC). Based on histopathological assessment, patients were divided into two groups: good responders (n = 57) and poor responders (n = 95). Significant differences in change rates for tumor diameter and volume were observed between the two groups (P < 0.001). The Delta-Radiomics model demonstrated superior predictive performance compared to other models, achieving an AUC of 0.796, ACC of 0.756, sensitivity of 0.529, specificity of 0.893, PPV of 0.750, and NPV of 0.758 in the test set. However, the Delong test revealed no significant differences among these models, except between the Post-NACT and Delta-Radiomics models (P < 0.05). T2-based MRI delta-radiomics showed strong predictive value for NACT response in patients with osteosarcoma. This model holds potential for guiding clinical decision-making and improving patient management by identifying responders early in the treatment course.

Background on statistical methods applied and the results from the bone only cohort analysis.

PURPOSE:Contemporary prostate cancer prognostic models do not include imaging and generally are based on pretreatment parameters. We sought to develop an externally validated model that used novel quantification of soft-tissue and bone disease, integrated with standard clinical and serum biomarkers, at baseline and up to 6 months of treatment. EXPERIMENTAL DESIGN:Two randomized phase 3 trials, Cougar COU-AA-302 (NCT00887198; for derivation) and Alliance A031201 (NCT01949337; for validation), were used to evaluate the added value of early on-treatment bone imaging and more than 1,000 radiomics features on CT, used in conjunction with clinical and serum biomarkers in first-line metastatic castration-resistant prostate cancer. Predictive accuracy measures were computed to determine whether these early on-treatment biomarkers could reliably sort patients into risk groups that inform overall survival (OS) and whether the patient-specific biomarker risk score could precisely predict their OS time. RESULTS:Imaging improved patient risk stratification but did not improve individual survival predictions. The strongest risk prediction model was developed for patients with bone-only metastases. This model was also the least complex, relying on just 16 risk factors, whereas all other models were high-dimensional, incorporating approximately 1,100 intercept and 1,100 slope features from the early on-treatment biomarker trajectories. CONCLUSIONS:Pretreatment and early on-treatment serum and automated quantitative imaging markers can well discriminate risk of death. Imaging improves this risk categorization relative to serum biomarkers alone. Such models can give early outcome predictions and can be used in future trials that involve imaging, even using traditional techniques such as bone scintigraphy.

Background/Objectives: Inter-reader variation can alter unidimensional tumor measurements used in RECIST-oriented response assessment. Unlike systems that automate segmentation or target selection, this study aimed to predict lesion-specific measurement uncertainty itself. Methods: The development cohort comprised 463 lung, liver, and lymph-node lesions from 280 patients in Vol-PACT, with four segmentations per lesion. Inter-reader variability was defined as (maximum longest diameter-minimum longest diameter)/minimum longest diameter. A Swin Transformer was trained to regress this continuous score. A value > 0.20 was used as a pragmatic high-variability alert threshold, with high variability designated as the positive class; this threshold is not equivalent to RECIST progressive disease. Preliminary external evaluation used 21 NSCLC lesions with five reader contours, providing one reader mask at a time. Results: The validation and internal test AUCs were 0.74 (95% CI, 0.61-0.85) and 0.89 (95% CI, 0.80-0.95), respectively. On the internal test set, MAE was 0.10 (95% CI, 0.08-0.12), accuracy was 82% (95% CI, 74-90%), sensitivity 72% (95% CI, 59-85%), and specificity for lower-variability lesions was 92% (95% CI, 82-100%). Across the five external readers, AUCs ranged from 0.77 to 0.92; the wide confidence intervals reflect the small external sample. Conclusions: This proof-of-concept model may provide a measurement-quality signal that supports target selection or adjudication. Larger multi-institutional studies, patient-level resampling, calibration, ablation testing, and workflow validation are required before clinical use.

Background Progression-free survival and objective response rate assessed per Response Evaluation Criteria in Solid Tumors, version 1.1 (RECIST v1.1), are based on serial radiographic assessments of tumor burden. We quantified case-level discordance between investigator and blinded independent central review (BICR)-determined progressive disease (PD) and objective response (OR) to provide benchmarks that can help clinical study teams judge whether observed discordance rates are expected or a data quality concern. Methods We performed a pooled analysis of 39 phase 2 and 3 solid-tumor trials of pembrolizumab-based regimens in which radiographic images were reviewed by investigators and BICR and RECIST v1.1 was used to assess response. Case-level discordance of PD, confirmed OR, and unconfirmed OR were calculated from 2×2 tables. Results Case-level discordance was 17.7% (95% CI 17.2–18.2) for PD (n=20,908), 12.2% (11.8–12.7) for confirmed OR (n=21,520), and 14.7% (14.1–15.3) for unconfirmed OR (n=15,209). Discordance varied by cancer type, ranging from 11.8% (melanoma) to 22.8% (hepatocellular carcinoma [HCC]) for PD, from 5.4% (HCC) to 20.0% (cervical cancer) for confirmed OR, and from 6.3% (HCC) to 19.7% (cervical cancer) for unconfirmed OR. Trials with real-time verification of progression showed lower PD discordance than those without (17.3% [95% CI 16.7-17.8] vs 20.2% [18.8-21.7]). PD discordance was similar in double-blind and open-label trials (17.6% [95% CI 16.9-18.3] vs 17.8% [17.0-18.6]). Confirmed OR discordance was more common in double-blind trials (13.3% [12.7-13.9] vs 10.9% [10.3-11.6]). Conclusion This analysis of trials of pembrolizumab-based regimens addresses a critical gap in oncology trial methodology by establishing empirical benchmarks for site–central discordance of RECIST v1.1 response assessment, enabling study teams to distinguish expected variability from potential data quality signals. While the underlying sources of discordance characterized here are not specific to immunotherapy, the reported magnitude of discordance should be applied cautiously to trials of non-immunotherapy agents.

BACKGROUND:Radiomics can provide quantitative descriptors of tumor phenotype, but translation is often limited by feature instability across scanners and protocols. We aimed to develop and internally validate a protocol-specific CT-radiomics model using preoperative imaging to predict 5-year recurrence in patients with stage I lung adenocarcinoma after complete surgical resection. METHODS:The retrospective study included 270 patients with completely resected stage I lung adenocarcinoma from January 2010-December 2021, among whom 23 (8.5%) experienced recurrence within five years. Radiomic features were extracted from routine preoperative CT scans. After preprocessing to remove highly constant and highly correlated features, the Synthetic Minority Over-sampling Technique addressed class imbalance in the training set. Recursive Feature Elimination identified the most predictive radiomic features. An XGBoost classifier was trained using optimized hyperparameters identified through RandomizedSearchCV with cross-validation. Model performance was evaluated using the ROC curve and predictive metrics. RESULTS:Five radiomic features differed significantly between recurrence groups (p = 0.007 to <0.001): Shape Sphericity, first-order 90Percentile, GLCM Autocorrelation, GLCM Cluster Shade, and GLDM Large Dependence Low Gray Level Emphasis. The radiomics model showed excellent discriminatory ability with AUC values of 0.99 (95% CI: 0.98-1.00), 0.97 (95% CI: 0.91-1.00), and 0.96 (95% CI: 0.85-1.00) on the training, validation, and test sets, respectively. On the test set, the model achieved sensitivity of 100% (95% CI: 51-100%), specificity of 94% (95% CI: 81-98%), PPV of 67% (95% CI: 30-90%), NPV of 100% (95% CI: 90-100%), and overall accuracy of 95% (95% CI: 83-99%). CONCLUSIONS:Under protocol-homogeneous imaging conditions, CT radiomics accurately predicted recurrence in patients with completely resected stage I lung adenocarcinoma. External multi-vendor validation is needed before broader deployment.

Background Real-time methods are needed for intraprocedural detection of residual tumors and incomplete thermal ablation (TA) to allow immediate retreatment and tumor eradication. Purpose To validate a TA workflow for detecting and immediately ablating residual viable colorectal liver metastases (CLMs). Materials and Methods This prospective single-center trial enrolled participants who underwent PET/CT-guided microwave CLM ablation from November 2019 to February 2023. The minimal ablation margin (MM) was calculated in all directions. Biopsies were obtained from the ablation zone (AZ) center and margin, with rapid tissue assessment for viable tumor (VT) cells using imprint cytology and fluorescent viability staining. Immediate reablation was performed if any of the following criteria were met: MM less than 5 mm at contrast-enhanced CT, residual PET-avid tumor, and/or VT cells at rapid tissue assessment. Gray-model statistics quantified the MM and VT impact on local tumor progression subdistribution hazard amid the competing risk of death. Results Seventy-seven participants (median age, 56 years [IQR, 47-64.5 years]; 39 male participants) underwent ablation in 104 CLMs. Overall, 15 of 104 (14%) CLMs underwent immediate reablation per the criteria (12 of 15, VT; seven of 15, MM <5 mm; and four of 15, residual fluorodeoxyglucose avidity). After reablation, all 12 initially VT-positive AZs underwent repeat biopsies with negative findings. Five of seven MMs less than 5 mm in AZs increased to greater than 5 mm after reablation. All four CLMs that underwent reablation due to PET/CT findings had AZs positive for VT, and one had MM less than 5 mm. MM greater than 5 mm protected against local tumor progression (LTP) (subdistribution hazard ratio, 0.12; 95% CI: 0.05, 0.30; P < .001). There was no LTP for MMs greater than 10 mm. The cumulative LTP incidence at 1, 2, and 3 years for participants with biopsy-proven completely ablated CLMs with MM greater than 5 mm was 7%, 12%, and 12%, respectively. Conclusion MM remained a critical technical factor affecting tumor control; the proposed multimodal comprehensive AZ assessment enabled immediate onsite reablation of 14% of CLMs with initially insufficient ablation treatment and improved local tumor control after thermal ablation. ClinicalTrials.gov identifier: NCT04143516 © RSNA, 2026 Supplemental material is available for this article. See also the editorial by Georgiades in this issue.

To develop and evaluate an automated CT liver lesion-tracking algorithm that matches lesions over time, detects new metastases, and reports per‑lesion confidence to support response assessment. The study included 87 adults with unresectable colorectal liver metastases (CRLM) who had baseline and 8-week follow-up contrast-enhanced CT. Three radiologists generated a consensus reference. We developed a machine learning-driven, automated model-based lesion tracking (Auto-MBT) that provides per-lesion matching confidence. Performance was compared with: (1) deformable registration + overlap; (2) deformable registration + Auto-MBT; and (3) affine registration + Auto-MBT. Analyses were stratified by lesion size (< 1 cm, 1–3 cm, overall) and count (≤ 5, 6–10, > 10 per scan), and the triage utility of confidence scores was assessed by blinded adjudication. A publicly available melanoma dataset was used for external testing. On the CRLM test set (35 pairs), affine + Auto-MBT matched 458/464 lesions (precision/recall 99

Importance In a phase 2 randomized clinical trial, high-dose vitamin D 3 added to standard treatment improved progression-free survival (PFS) compared with standard-dose vitamin D 3 in patients with metastatic colorectal cancer (mCRC). Objective To determine if high-dose vitamin D 3 added to standard chemotherapy improves outcomes in patients with previously untreated mCRC. Design, Setting, and Participants Double-blind phase 3 randomized clinical trial enrolling 455 patients with previously untreated mCRC, conducted in the US through the National Clinical Trials Network from October 2019 to December 2022 (database freeze: July 15, 2024). Interventions mFOLFOX6 (modified FOLFOX6 [5-fluorouracil, leucovorin, oxaliplatin]) or FOLFIRI (5-fluorouracil, leucovorin, irinotecan) plus bevacizumab every 2 weeks with either high-dose vitamin D 3 (8000 IU daily × 14 days as loading dose followed by 4000 IU daily) or standard-dose vitamin D 3 (400 IU daily) until disease progression, intolerable toxicity, or withdrawal of consent. Main Outcomes and Measures The primary end point was PFS assessed by the unstratified log-rank test. Secondary end points included objective response rate, overall survival, and toxicity. Prespecified subgroup analyses of PFS were performed according to known prognostic factors. Results Among 455 randomized patients (median age, 59 years; 181 [40%] female) with median follow-up 20 months, the median PFS for high-dose vitamin D 3 (n = 228) was 11.8 months (95% CI, 10.3-13.3) vs 10.3 months (95% CI, 9.4-12.2) for standard-dose vitamin D 3 (n = 227) (1-sided log-rank P = .25). There were no significant differences in objective response rate between high-dose and standard-dose vitamin D 3 (51% [95% CI, 44%-58%] vs 44% [95% CI, 37%-50%], respectively; P = .12), or in overall survival (median, 25.6 vs 27.0 months; 1-sided log-rank P = .66). There were no clinically meaningful differences in the most common grade 3 or greater adverse events between the high- and standard-dose groups, including neutropenia (n = 67 [32%] vs n = 62 [30%]) and hypertension (n = 42 [20%] vs n = 49 [23%]) or in incidence of vitamin D–associated toxicities. Conclusions and Relevance Among patients with previously untreated mCRC, addition of high-dose vitamin D 3 , vs standard-dose vitamin D 3 , to standard chemotherapy plus bevacizumab did not improve PFS. Trial Registration ClinicalTrials.gov Identifier: NCT04094688

BACKGROUND AND OBJECTIVE:Neoadjuvant immune-checkpoint inhibitors (ICIs) in muscle-invasive bladder cancer (MIBC) were tested in patient's ineligible for cisplatin-based chemotherapy. The PURE-01 trial (NCT02736266) evaluated three courses of pembrolizumab before radical cystectomy (RC). We developed AI-MIRACLE, an international study assessing artificial intelligence (AI) and multiparametric magnetic resonance imaging (mpMRI) for predicting treatment response. METHODS:This multi-institutional study included data acquisition in Italy, and centralized analysis in the United States. Among 112 PURE-01 patients, pre- and post-ICI MRIs were analyzed. T2-weighted signal intensities were standardized for radiomics (Image Biomarker Standardization Initiative-compatible Python-based Computational Environment for Radiological Research (pyCERR)) and deep feature extraction (AI-BLADE toolbox using VGG19). Diffusion-weighted (DW) and dynamic contrast-enhanced (DCE) MRI data underwent model-based analysis. Supervised machine learning algorithms (elastic net, random forest) were trained and cross-validated to predict pathological major response (pMR:ypT<2N0 residual disease) and pathological complete response (pCR: ypT0) pathological response. KEY FINDINGS AND LIMITATIONS:The predictive models using post-ICI mpMRI with either a combination of radiomics and DCE-derived features or radiomics alone achieved the same high accuracy, with an area under the receiver operating characteristic curve (AUC) of 0.96 for pMR. A shape-based radiomic model achieved an AUC of 0.86 for predicting pCR. These models outperformed benchmark models based on clinical predictors. CONCLUSIONS AND CLINICAL IMPLICATIONS:Shape-based radiomics, DCE-derived features, and deep features may serve as noninvasive imaging biomarkers for predicting response to neoadjuvant pembrolizumab in MIBC. This imaging-based approach provides a non-invasive assessment of treatment response following neoadjuvant immunotherapy, which may help inform bladder-preserving management decisions prior to definitive surgery.

Chimeric antigen receptor (CAR) T-cell therapy has revolutionized the treatment landscape for multiple hematologic malignancies, demonstrating a clear curative potential in select cases. Since 2017, 7 CAR T-cell products have indications approved by the Food and Drug Administration, and research is continuously evolving with the intent of enhancing CAR T-cell efficacy while reducing side effects and the risk of relapse. Nuclear medicine has played a crucial role throughout this optimization process, assisting initial staging, treatment response assessment, and adverse-effect monitoring. This review provides an introduction to CAR T-cell therapy and highlights gold-standard nuclear medicine practices and imaging innovations for use in the care of oncology patients. Additionally, it emphasizes the need for continuing innovation in medical imaging as the field progresses.

Background and objective Accurate bladder tumor segmentation is crucial for muscle invasion assessment, neoadjuvant therapy response evaluation, and bladder preservation decision‑making. Manual delineation is time‑consuming and observer‑dependent, highlighting an unmet clinical need for accurate and automated segmentation, particularly in large multi‑site studies. This study aims to (i) develop an automated MRI‑based foundation model for bladder tumor segmentation (BLA‑T‑Seg) using a two‑stage framework, in which segmentation is first constrained to the bladder wall, including wall‑contiguous tumors and then refined in a second stage focused specifically on the tumor, and (ii) rigorously assess generalizability via multi-site, independent external validation across diverse cohorts. Methods This retrospective study included T2‑weighted MRI scans from 236 patients with bladder cancer across two international sites. Experienced radiologists manually annotated bladder and tumor regions of interest on T2‑weighted images, which served as the ground truth. BLA‑T‑Seg was developed by adapting the Segment Anything Model (SAM), fine‑tuned to segment the bladder in stage 1 and the tumor in stage 2. The primary performance metric was the Dice similarity coefficient (DSC), evaluated across five studies (S1–S3: internal; S4–S5: external) and benchmarked against 10 state‑of‑the‑art (SOTA) segmentation architectures. Results For stage 1, BLA‑T‑Seg achieved good‑to‑excellent performance, with mean DSC values ranging from 0.85 to 0.93. For stage 2, BLA‑T‑Seg demonstrated moderate‑to‑good performance, with mean DSC values of 0.74–0.81. Cross‑site validation showed minimal performance variation (≤0.01–0.04), confirmed by radar plot analyses. Compared with SOTA models on subset S1, BLA‑T‑Seg achieved the highest DSC of 0.84, outperforming the next best‑performing model, Swin Transformer (DSC: 0.56). Conclusion BLA‑T‑Seg enables accurate and generalizable bladder wall and tumor segmentation and outperforms SOTA alternatives.

PURPOSE:The continuous development of new imaging approaches, molecular phenotyping, genetic subtypes, prognosis assessments, and effective therapies across a range of disease states has created a need to redefine terminology and best practices for clinical trial conduct in patients with advanced prostate cancer. METHODS:We convened an international expert committee of diverse working groups, the Prostate Cancer Working Group 4 (PCWG4), between 2016 and 2025. Our objective was to formulate updated criteria based on emerging evidence and clinical trial data in a biomarker context to provide guidance for clinical trial design, eligibility, and end point assessments for patients with advanced prostate cancer. RESULTS:PCWG4 redefines terminology around the disease state and previous therapies in a patient-centric context and terminology focused on androgen pathway modulation. We consider imaging, with a particular focus on positron emission tomography (PET)-defined disease. New recommendations are provided for disease state terminology, defining eligibility criteria, response and delay/prevent end points, intervals for reassessments including imaging, and patient-reported outcome determination. We provide recommendations in a biomarker-based context of use for the intended indication, reflective of patient benefit for specific interventions. We emphasize the need for development of validated PET imaging and molecular and phenotypic criteria as well as trial designs to appropriately risk stratify patients, predict and assess benefit, and measure post-treatment outcomes reliably in a trial framework. CONCLUSION:PCWG4 updates recommendations on patient and tumor characterization, therapy development, and imaging criteria and extends guidance into earlier androgen pathway modulator-naïve/sensitive disease states to reflect an evolving, heterogeneous, and diverse patient population to optimize treatment benefits for all patients.

Background/Objectives: The tumor microenvironment (TME) of pancreatic ductal adenocarcinoma (PDAC) is characterized by an enriched stroma, hampering the effectiveness of therapy. This co-clinical study aimed to (1) provide insight into early post-treatment changes in the TME using multiparametric magnetic resonance imaging (mpMRI)-derived quantitative imaging biomarkers (QIBs) in a preclinical PDAC model treated with radiotherapy and correlate these QIBs with histology; (2) evaluate the feasibility of obtaining these QIBs in patients with PDAC using clinically approved mpMRI data acquisitions. Methods: Athymic mice (n = 12) at pre- and post-treatment as well as patients with PDAC (n = 11) at pre-treatment underwent mpMRI including diffusion-weighted (DW) and dynamic contrast-enhanced (DCE) data acquisition sequences. DW and DCE data were analyzed using monoexponential and extended Tofts models, respectively. DeepLIIF quantified the total percentage (%) of tumor cells in hematoxylin and eosin (H&E)-stained tissues from athymic mice. Spearman correlation and Wilcoxon signed rank tests were performed for statistical analysis. Results: In the preclinical PDAC model, mean pre- and post-treatment ADC and Ktrans values differed significantly (p < 0.01), changing by 20.50% and 20.41%, respectively, and the median total tumor cells quantified by DeepLIIF was 24% (range: 15-53%). Post-treatment ADC values and relative change in ve (rΔve) showed a significant negative correlation with total tumor cells (ρ = -0.77, p < 0.014 for ADC and ρ = -0.77, p = 0.009 for rΔve). In patients with PDAC, pre-treatment mean ADC and Ktrans values were 1.76 × 10-3 (mm2/s) and 0.24 (min-1), respectively. Conclusions: QIBs in both preclinical and clinical settings underscore their potential for future co-clinical research to evaluate emerging drug combinations targeting both tumor and stroma.

264 Background: In mCRPC, standard risk models use baseline clinical and blood-based biomarkers, excluding imaging. This study developed a new predictive model incorporating baseline and early on-treatment clinical, blood, and imaging biomarkers to inform overall survival (OS). The training set was derived from COU-AA-302 and validated using Alliance 031201 (A031201). Eligibility required patients to have at least two CT and bone scans within 6 months of randomization (COU-AA-302: n = 785; A031201: n = 582). Methods: Conventional biomarker data included PSA, albumin, alkaline phosphatase, hemoglobin, and ECOG status. Imaging data included > 1000 radiomic features and the automated Bone Scan Index (aBSI). Six-month biomarker trajectories were summarized using the intercept and slope from a random effects model, yielding over 2000 risk biomarkers. Due to the high dimensionality, an elastic net proportional hazards model was applied. Predictive concordance probability, which evaluates the ability to distinguish between long-term and short-term survivors, and R-square, a calibration measure that indicates how accurately the model predicts actual survival time were determined. Results: 46 biomarkers were selected from over 2000, including 9 blood-based, 4 from aBSI, 1 tumor volume measure, and the remaining from radiomic features. In contrast, the non-imaging risk model contained 10 risk factors. Risk scores from these models were categorized into low, moderate, and high-risk groups. The median survival per group revealed significant separation, with a difference of over 30 months between low and high-risk groups. This separation was observed regardless of whether imaging was included in the model. Conclusions: This is the first model for mCRPC to integrate pre-treatment and early on-treatment serum, imaging, and clinical data from two large phase 3 trials, with external validation. While both imaging and non-imaging models showed good predictive accuracy, imaging did not significantly improve the discrimination of long-term from short-term survivors or enhance the precision of the model-based survival predictions. Clinical + Serum biomarkers with imaging Clinical + Serum biomarkers without imaging Total # of risk factors 46 10 # of clinical/serum factors 9 10 # of radiomics factors 32 0 # of aBSI factors 4 0 # of tumor volume factors 1 0 Median Survival (mos) Low risk group 55 54 Moderate risk group 38 39 High risk group 20 21 Predictive Concordance Probability 0.74 0.73 R-square Measure of Calibration 0.37 0.39

Radiologic nodal staging in CTCL traditionally uses a 1.5 cm longest diameter (LDi) cutoff; however, this lacks validation and may misclassify risk. We conducted a retrospective analysis of 6,095 CT scans from 262 CTCL patients in the MAVORIC trial using unidimensional, bidimensional, and volumetric LN measurements and mSWAT scores. Optimal cutoffs were determined via ROC analysis and landmarking adjusted for informative censoring. Additionally, kinetic modeling growth rates (g) were calculated for both LN and skin scores. We demonstrated that LDi > 1.5 cm did not predict OS (p = 0.8). However, baseline volumetric cutoffs (3,945 mm3; AUC = 0.67) stratified OS (median 43.6 months vs NA, log-rank p = 0.035); post‑landmark analysis (6,930 mm3) enhanced discrimination. High g (volumetric or mSWAT) independently predicted shorter OS/PFS/TTF (p < 0.05). A combined model (volume + g) had C‑index 0.63 versus 0.60 for volume alone. We conclude that volumetric and dynamic metrics outperform conventional measures in predicting CTCL outcomes. Incorporating these methods into staging and trial criteria is warranted.

Objectives: Accurate kidney and tumor segmentation of computed tomography (CT) scans is vital for diagnosis and treatment, but manual methods are time-consuming and inconsistent, highlighting the value of AI automation. This study develops a fully automated AI model using vision transformers (ViTs) and convolutional neural networks (CNNs) to detect and segment kidneys and kidney tumors in Contrast-Enhanced (CECT) scans, with a focus on improving sensitivity for small, indistinct tumors. Methods: The segmentation framework employs a ViT-based model for the kidney organ, followed by a 3D UNet model with enhanced connections and attention mechanisms for tumor detection and segmentation. Two CECT datasets were used: a public dataset (KiTS23: 489 scans) and a private institutional dataset (Private: 592 scans). The AI model was trained on 389 public scans, with validation performed on the remaining 100 scans and external validation performed on all 592 private scans. Tumors were categorized by TNM staging as small (≤4 cm) (KiTS23: 54%, Private: 41%), medium (>4 cm to ≤7 cm) (KiTS23: 24%, Private: 35%), and large (>7 cm) (KiTS23: 22%, Private: 24%) for detailed evaluation. Results: Kidney and kidney tumor segmentations were evaluated against manual annotations as the reference standard. The model achieved a Dice score of 0.97 ± 0.02 for kidney organ segmentation. For tumor detection and segmentation on the KiTS23 dataset, the sensitivities and average false-positive rates per patient were as follows: 0.90 and 0.23 for small tumors, 1.0 and 0.08 for medium tumors, and 0.96 and 0.04 for large tumors. The corresponding Dice scores were 0.84 ± 0.11, 0.89 ± 0.07, and 0.91 ± 0.06, respectively. External validation on the private data confirmed the model’s effectiveness, achieving the following sensitivities and average false-positive rates per patient: 0.89 and 0.15 for small tumors, 0.99 and 0.03 for medium tumors, and 1.0 and 0.01 for large tumors. The corresponding Dice scores were 0.84 ± 0.08, 0.89 ± 0.08, and 0.92 ± 0.06. Conclusions: The proposed model demonstrates consistent and robust performance in segmenting kidneys and kidney tumors of various sizes, with effective generalization to unseen data. This underscores the model’s significant potential for clinical integration, offering enhanced diagnostic precision and reliability in radiological assessments.